hlist_add_tail_rcu disable sparse warning
PATCH 1/1 hlist_add_tail_rcu disable sparse warning
sparse is unhappy about this code in hlist_add_tail_rcu:
struct hlist_node *i, *last = NULL;
for (i = hlist_first_rcu(h); i; i = hlist_next_rcu(i))
last = i;This is because hlist_next_rcu and hlist_next_rcu return __rcu pointers.
It's a false positive - it's a write side primitive and so does not need to be called in a read side critical section.
The following trivial patch disables the warning
without changing the behaviour in any way.
Note: __hlist_for_each_rcu would also remove the warning but it would be confusing since it calls rcu_derefence and is designed to run in the rcu read side critical section.
---
changes since RFC
added commit log text to explain why don't we use __hlist_for_each_rcu
include/linux/rculist.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/rculist.h +1 -1
@@ -509,7 +509,7 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n, {struct hlist_node *i, *last =NULL ; -for (i =hlist_first_rcu (h); i; i =hlist_next_rcu (i)) +for (i = h->first; i; i = i->next) last = i;if (last) {
Signature · 1 line
MST
include/linux/rculist.h static inline void hlist_add_tail_rcu(struct hlist_node *n, · line 512
Quoted 10 lines · On Fri, 10 Feb 2017 19:39:49 +0200 "Michael S. Tsirkin" <[email protected]> wrote:
include/linux/rculist.h +1
-1@@ -509,7 +509,7 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n, {struct hlist_node *i, *last =NULL ; -for (i =hlist_first_rcu (h); i; i =hlist_next_rcu (i)) +for (i = h->first; i; i = i->next)
Looks good to me, but I would probably add a comment above the loop about this being in a write side section, and rcu conversions are not needed.
-- Steve
Quoted 3 lines
include/linux/rculist.h +0
-0last = i;if (last) {
Quoted 16 lines · Michael S. Tsirkin
sparse is unhappy about this code in
hlist_add_tail_rcu:struct hlist_node *i, *last = NULL; for (i = hlist_first_rcu(h); i; i = hlist_next_rcu(i)) last = i;This is because
hlist_next_rcuandhlist_next_rcureturn__rcupointers.It's a false positive - it's a write side primitive and so does not need to be called in a read side critical section.
The following trivial patch disables the warning
without changing the behaviour in any way.
Quoted 6 lines
Note:
__hlist_for_each_rcuwould also remove the warning but it would be confusing since it callsrcu_derefenceand is designed to run in the rcu read side critical section.Signed-off-by Michael S. Tsirkin <[email protected]>---
ping
Quoted 21 lines
changes since RFC
added commit log text to explain why don't we use__hlist_for_each_rcuinclude/linux/rculist.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)include/linux/rculist.h +1
-1@@ -509,7 +509,7 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n, {struct hlist_node *i, *last =NULL ; -for (i =hlist_first_rcu (h); i; i =hlist_next_rcu (i)) +for (i = h->first; i; i = i->next) last = i;if (last) {Signature · 1 line
MST
Steven Rostedt
reviewerPaul,
Did you see this email?
-- Steve
Quoted 47 lines · On Mon, 27 Feb 2017 20:26:01 +0200 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 16 lines · Michael S. Tsirkin
sparse is unhappy about this code in
hlist_add_tail_rcu:struct hlist_node *i, *last = NULL; for (i = hlist_first_rcu(h); i; i = hlist_next_rcu(i)) last = i;This is because
hlist_next_rcuandhlist_next_rcureturn__rcupointers.It's a false positive - it's a write side primitive and so does not need to be called in a read side critical section.
The following trivial patch disables the warning
without changing the behaviour in any way.Quoted 6 lines
Note:
__hlist_for_each_rcuwould also remove the warning but it would be confusing since it callsrcu_derefenceand is designed to run in the rcu read side critical section.Signed-off-by Michael S. Tsirkin <[email protected]>---
ping
Quoted 21 lines
changes since RFC
added commit log text to explain why don't we use__hlist_for_each_rcuinclude/linux/rculist.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)include/linux/rculist.h +1
-1@@ -509,7 +509,7 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n, {struct hlist_node *i, *last =NULL ; -for (i =hlist_first_rcu (h); i; i =hlist_next_rcu (i)) +for (i = h->first; i; i = i->next) last = i;if (last) {Signature · 1 line
MST
Quoted 4 lines · Steven Rostedt
Paul,
Did you see this email?
Yep! Michael's patch is 48ac34666ff7 ("hlist_add_tail_rcu disable sparse warning") in -rcu and now in -tip.
But I do appreciate the reminder -- way too easy to miss stuff on LKML!
Thanx, Paul
Quoted 54 lines
-- Steve
Quoted 47 lines · On Mon, 27 Feb 2017 20:26:01 +0200 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 16 lines · Michael S. Tsirkin
sparse is unhappy about this code in
hlist_add_tail_rcu:struct hlist_node *i, *last = NULL; for (i = hlist_first_rcu(h); i; i = hlist_next_rcu(i)) last = i;This is because
hlist_next_rcuandhlist_next_rcureturn__rcupointers.It's a false positive - it's a write side primitive and so does not need to be called in a read side critical section.
The following trivial patch disables the warning
without changing the behaviour in any way.Quoted 6 lines
Note:
__hlist_for_each_rcuwould also remove the warning but it would be confusing since it callsrcu_derefenceand is designed to run in the rcu read side critical section.Signed-off-by Michael S. Tsirkin <[email protected]>---
ping
Quoted 21 lines
changes since RFC
added commit log text to explain why don't we use__hlist_for_each_rcuinclude/linux/rculist.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)include/linux/rculist.h +1
-1@@ -509,7 +509,7 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n, {struct hlist_node *i, *last =NULL ; -for (i =hlist_first_rcu (h); i; i =hlist_next_rcu (i)) +for (i = h->first; i; i = i->next) last = i;if (last) {Signature · 1 line
MST